Subscriber access provided by Duquesne University - Gumberg Library
Article
In situ Solid-state 13C NMR Observation of Pore Mouth Catalysis in Etherification of #-Citronellene with Ethanol on Zeolite Beta Sambhu Radhakrishnan, Pieter-Jan Goossens, Pieter C.M.M. Magusin, Sreeprasanth Pulinthanathu Sree, Christophe Detavernier, Eric Breynaert, Charlotte Martineau, Francis Taulelle, and Johan A. Martens J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.5b13282 • Publication Date (Web): 04 Feb 2016 Downloaded from http://pubs.acs.org on February 6, 2016
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 8
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Journal of the American Chemical Society
In situ Solid-state 13C NMR Observation of Pore Mouth Catalysis in Etherification of β-Citronellene with Ethanol on Zeolite Beta Sambhu Radhakrishnan†, Pieter-Jan Goossens†, Pieter C. M. M. Magusin†, Sreeprasanth Pulinthanathu Sree†, Christophe Detavernier‡, Eric Breynaert†, Charlotte Martineau¥,#, Francis Taulelle†,¥ and Johan A. Martens†,* † Centre for Surface Chemistry and Catalysis, KU Leuven, Celestijnenlaan 200F Box 2461, 3001-Heverlee, Leuven, Belgium. ¥ Tectospin, Institut Lavoisier, UMR 8180, Université de Versailles st. Quentin en Yvelines 45, Avenue des Etats Unis, 78035 Versailles, Cedex, France. # CEMHTI, CNRS UPR 3079, 1D Avenue de la Recherche Scientifique, 45071 Orléans Cedex 2, France ‡ Department of Solid State Sciences, CoCooN, Ghent University, Krijgslaan 281 (S1), Ghent 9000, Belgium KEYWORDS: Zeolite beta, Etherification, Pore mouth catalysis, Citronellene, In situ NMR
ABSTRACT: The reaction mechanism of etherification of β-citronellene with ethanol in liquid phase over acid zeolite beta is revealed by in situ solid-state 13C NMR spectroscopy. Comparison of 13C Hahn-echo and 1H – 13C cross polarization NMR characteristics are used to discriminate between molecules freely moving in liquid phase outside the zeolite and molecules adsorbed inside zeolite pores and in pore mouths. In the absence of ethanol, β-citronellene molecules enter zeolite pores and react to isomers. In the presence of ethanol, the concentration of β-citronellene inside zeolite pores is very low because of preferential adsorption of ethanol. The etherification reaction proceeds by adsorption of β-citronellene molecule from the external liquid phase in a pore opening where it reacts with ethanol from inside the pore. By competitive adsorption, ethanol prevents the undesired side reaction of βcitronellene isomerization inside zeolite pores. β-citronellene etherification on zeolite beta is suppressed by bulky base molecules (2,4,6-collidine and 2,6-ditertiarybutyl pyridine) that do not enter the zeolite pores confirming the involvement of easily accessible acid sites in pore openings. The use of in situ solid-state NMR to probe the transition from intracrystalline catalysis to pore mouth catalysis depending on reaction conditions is demonstrated for the first time. The study further highlights the potential of this NMR approaches for investigations of adsorption of multicomponent mixtures in general. Introduction Catalytic processes on zeolites generally involve diffusion and reaction of molecules throughout the intracrystalline pore system.1 Bulky molecules unable to penetrate the pores can only react in partial cavities and pore openings at zeolite crystal boundaries offering more space than deeper inside the pores.2 Pore blockage by coke deposition inside the pores can be another reason why the catalytic activity of a zeolite is located at the pore entrances.3 Pore mouth catalysis accordingly is a particular mechanism for specific combinations of reagent and zeolite. Venuto has been the first to propose pore mouth catalysis in aromatic alkylation.4 The notion of pore mouth catalysis has been elaborated later on by Fraenkel5 and Derouane6 who pointed out that the pockets on the zeolite surface formed by pore termination and cut pore intersections provide a special steric environment for reaction. The peculiar methyl branching selectivity of long n-alkanes on medium-pore zeolites such as ZSM-22 has been explained by pore mouth catalysis.7,8 Long molecules able to span the distance between two neighboring pore openings can penetrate simultaneously with both ends in a separate pore and react according to a key-lock principle.9,10 In hydrocracking on zeolite Y being a large-pore zeolite, the penetration of heavy alkanes in zeolite crystals seems to be limited to a few nanometers only.11 Direct exper-
imental evidence for pore mouth catalysis is however difficult to provide. Experimental evidence has been based on adsorption studies,12 estimation of molecular diffusivity13 and molecular and kinetic modelling.14,15 Zeolite beta is a prominent catalyst for many applications, and especially for processes related to biomass conversion.16–22 Sn-beta zeolite catalyst is popular in green chemistry. Examples of catalyzed reactions are conversion of 1,3dihydroxy acetone to ethyl lactate,20 isomerization of glucose in aqueous solutions,21 and Baeyer-Villiger oxidation reactions.22 Recently we showed acid zeolite beta catalyzes etherification of linear α-olefins and β-citronellene with alcohols.23– 25 In many reactions in the liquid phase, zeolite beta outperforms other zeolites. NMR is a versatile tool to probe zeolite catalytic chemistry in situ.26–35 There are few if any reports on the use of solid state NMR for in situ investigation of catalysis in mixed phase comprising a liquid reagent in contact with a zeolite. The etherification reaction of β-citronellene with ethanol to ethers proceeds under such reaction conditions at temperatures as low as 80°C.23 It is therefore a suitable model reaction for an in situ solid-state NMR investigation. Here we use the 13C NMR characteristics in Hahn-echo and 1H-13C cross polarization (CPMAS) spectra, to discriminate between molecules
ACS Paragon Plus Environment
Journal of the American Chemical Society
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
freely moving in liquid phase outside the zeolite and adsorbed molecules. From the ensemble of results, we propose an explanation for the exceptional catalytic performances of zeolite beta in a catalytic process in liquid phase.
Page 2 of 8
and ca. 2% 1-methyl-4-(propan-2-ylidene)cyclohexane isomers) were from Sigma Aldrich. n-Heptane was verified to be inert under reaction conditions and was used as internal standard. In catalyst poisoning experiments 1 wt% 2,4,6-collidine or 2,6-ditertiary butyl pyridine was added to the feed mixture of β-citronellene and ethanol. For GC analysis of reactions products using FID detector, response factors of identified reaction products were calculated by effective carbon number method.42
Experimental Ammonium zeolite beta (CP 814C Si/Al = 19) was obtained from Zeolyst. It was converted to the acid form by calcination at 450°C. Nitrogen adsorption isotherms at -196°C were determined on a Micromeritics Tristar 3000 instrument. Results and Discussion X-ray photoelectron spectra (XPS) were recorded using an SThe investigated zeolite beta sample is a spherical Probe Monochromatized XPS spectrometer from Surface Sciaggregate of crystallites measuring 20 to 60 nm (Figure 1). ence Instruments (VG) with monochromatic Al Kα X-ray The BET specific surface area is ca. 486 m2/g and the mi(1486.6 eV) source. Binding Energy values (BE) were correctcropore volume ca. 0.23 cm3/g, typical of zeolite beta.43 The ed for charging eff ects by assigning a BE of 284.6 eV to the presence of compositional gradients was probed by determining the Si/Al ratio at the external surface using XPS and comC1s signal. Scanning electron microscopy (SEM) was perparison with overall composition. The Si/Al atomic ratio of 17 formed on a FEI Nova NanoSEM450 (FEI) instrument at 2 at the external surface was similar to the overall Si/Al ratio of kV. FTIR spectra were recorded on a Nicolet 6700 spectrome19 revealing absence of significant compositional gradients. ter. A pressed self-supporting zeolite wafer weighing ca. 20 mg was mounted in a vacuum cell inside the spectrometer. The sample was evacuated at 400 °C for 1 h. Adsorption of pyridine and 2,4,6-collidine was performed by exposing the pellet to vapor of the compound (ca. 4 - 5 kPa) at 150°C for 10 minutes. Desorption was done by raising the temperature. Spectra were recorded at 200°C and 400°C. The molar extinction coefficient for Brønsted and Lewis acid site quantification using pyridine was taken from Emeis,36 and for adsorbed 2,4,6-collidinium ion from Nesterenko et al.37 (+)-β-citronellene (99%) used in the NMR studies Figure 1. Scanning electron micrographs of zeolite beta sample at was obtained from Sigma Aldrich and absolute ethanol from two magnifications. VWR. For in situ solid-state NMR spectroscopy acid zeolite Quantification of Brønsted and Lewis type acid sites beta powder was dried under N2 flow at 400 °C for 3 h. Dry was done by titration with pyridine and FTIR spectroscopy. zeolite powder was combined with β-citronellene or βThe pyridine molecule with kinetic diameter of 0.59 nm has citronellene – ethanol mixture (1:10 molar ratio) in proporaccess to zeolite beta pores having a minimum free diameter tions of 0.6 ml/g to obtain a wet powder. Contact with the of ca. 0.67 nm.44 Pyridine adsorbed on Brønsted and Lewis ambient was avoided to prevent uptake of air-borne water acid sites generates characteristic IR absorption bands at 1540 vapor. In situ MAS NMR experiments were performed on a cm-1 and 1450 cm-1, respectively.36,45 Brønsted acid sites are Bruker Ascend 500 MHz spectrometer (static magnetic field due to protons compensating negative framework charge defiof 11.7 T) with 4 mm H/X/Y MAS probe at room temperature cits caused by trivalent aluminum incorporation in tetrahedral 13 with a rotor spinning rate of 5 kHz. C NMR signal was framework. Lewis acid sites are typically due to extramonitored via two methods: direct excitation with Hahn-echo framework Al species. Zeolite beta has Brønsted acid sites acquisition38 and cross polarization (CP).39,40 For Hahn-echo mainly, and only low amounts of Lewis acid sites (Table 1). pulse sequence, a delay of 400 μs, two rotor periods, between Pyridine adsorption at 200°C was considered to probe all acid 90° and 180° pulses, , 4.5 μs 90° pulse length, recycle delay of sites. At 400°C, pyridine probes strong acid sites. A significant 3 s, 4096 scans, and 1H SPINAL-6441 decoupling (70 kHz) share of the Brønsted acid sites are strong (Table 1). Polymeduring signal acquisition were used. The 1H-13C CPMAS thyl substituted pyridine bases like 2,4,6-collidine (kinetic 13 NMR spectra were recorded using 50 kHz RF field on C, a diameter 0.74 nm)37,45 can be used for quantification of acid RAMPed pulse on 1H (100-80%) for the contact, 4096 scans, sites with less steric constraint than inside the pores. Thibaultrecycle delay of 3s, 1H SPINAL-64 decoupling41 and a contact Starzyk et al. proposed the accessibility index (ACI) as the time of 5 ms. Both experiments took ca. 3.5 hours. The 13C ratio of the number of acid sites detected by adsorption of the chemical shifts are referenced to TMS. probe to the total amount of acid sites in the zeolite, to quantiCatalytic etherification experiments were performed fy the share of acid sites with enhanced accessibility.45 For in a liquid phase, continuous flow, tubular reactor with interlarge pore zeolites like beta, all the acid sites will be accessible nal free diameter of 9 mm. The catalyst bed consisted of 1.5 g to pyridine and hence the ACI could be calculated as the ratio of compressed zeolite beta particles (diameter range of 250 – of the amount of acid sites accessible to bulky probe (2,4,6500 μm). A layer of quartz spheres (2 mm diameter) was posicollidine) to the amount of acid sites probed by pyridine.45 The tioned upstream for preventing channeling of the liquid flow. ACI of zeolite beta, determined using 2,4,6-collidine and pyriThe catalyst was pretreated at 200°C under N2 gas flow (100 dine for probing respectively Brønsted acid sites at pore ml/min) for 2 h followed by calcination at 450°C under O2 mouths and in total was 0.23. Zeolite beta has a three dimenflow (150 ml/min) for 2 h. n-heptane (>99%) and 2,4,6sional pore system with intersecting channels in the three dicollidine (>99%) were from Acros Organics; 2,6rections44 such that all crystal facets have pore openings. ditertiarybutyl pyridine (>97%) and β-citronellene (ca. 91% βBased on the ACI value, the zeolite beta crystal structure and citronellene, ca. 7% 1-isopropenyl-2,3-dimethyl-cyclopentane assuming uniform acid site concentration (absence of compoACS Paragon Plus Environment
Page 3 of 8
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Journal of the American Chemical Society
sitional gradients evidenced by XPS, vide supra), the penetration depth of 2,4,6-collidine into the zeolite crystallites measuring 40 nm on average was estimated at ca. 2.5 nm. This penetration depth is about thrice the molecular size of 2,4,6collidine. Despite the roughness of this estimation of penetration depth, it shows 2,4,6-collidine selectively probes acid sites at very limited depth. The ACI of zeolite beta is higher than for medium-pore zeolites like H-ZSM-5 (ACI = 0.06)45 and large-pore mordenite zeolite (ACI = 0.18).37 The small particle size (Figure 1) and the tridimensional pore system with pore openings on all crystal facets explains the high ACI value. Table 1. Acidity of zeolite beta Total acid sitesa (mmol/g)
Brønsted
0.516
Lewis
0.166
Strong acid sitesb (mmol/g)
Brønsted
0.230
Lewis
0.094
Total pore mouth sites (mmol/g)c
Brønsted
0.117
Strong pore mouth sites (mmol/g)d
Brønsted
0.078
Accessibility index (ACI)e
Brønsted
0.23
a, pyridine adsorption at 200°C; b, pyridine adsorption at 400°C; c, 2,4,6-collidine adsorption at 200°C; d, 2,4,6-collidine adsorption at 400°C; e, molar ratio of acid sites accessible to 2,4,6collidine to acid sites probed by pyridine at 200°C.
NMR investigations were performed using commercially available enantiopure (+)-β-citronellene liquid. Since enantioselectivity is beyond the scope of the present study, the compound further on is denoted simply as β-citronellene. NMR chemical shifts of 13C nuclei in β-citronellene, ethanol, the preferred etherification product (7-ethoxy-3,7-dimethyloct1-ene) and the hydration product 2,6-dimethyl-oct-7-en-2-ol were assigned via a combination of 1D and 2D liquid state NMR experiments (Figure 2).
Figure 2. Experimentally assigned 13C NMR chemical shifts (ppm) of (a) (+)-β-citronellene, (b) ethanol, (c) β-citronellyl ether - 7-ethoxy-3,7-dimethyloct-1-ene and (d) hydration product - 2,6-dimethyl-oct-7-en-2-ol. For this study, we rely on differences of relaxation properties of molecules inside and outside zeolite pores. Outside molecules are expected to be very mobile, liquid-like, leading to narrow NMR resonances and reduced 1H-13C residual dipolar couplings with very inefficient CP transfer and longitudinal relaxation times (T1) shorter than in the solidstate.46 On the contrary, molecules inside zeolite pores are expected to have reduced mobility and the possibility to adopt several conformations. This results in heterogeneous broadening of the 13C NMR resonances. NMR resonances for each chemically identical function will be distributed according to
the different conformations, each contributing a narrow component to the observed broad signal. 1H-13C residual dipolar coupling in adsorbed state is favorable for polarization transfer. Therefore, direct acquisition of 13C NMR spectra through a Hahn-echo sequence allows observation of both mobile molecules (narrow lines) and adsorbed molecules (broad signals) while 1H-13C CPMAS NMR spectra allow for selective observation of the adsorbed molecules. Line width reflects the degree of conformation distribution. To gain insight in adsorption and reaction of βcitronellene on zeolite beta, three different samples were investigated (Table 2). In Experiment 1, dehydrated zeolite beta was contacted with β-citronellene liquid. Experiment 2 was similar except zeolite beta was hydrated. In Experiment 3, dehydrated zeolite beta was contacted with a mixture of ethanol and β-citronellene. Table 2. Molar proportions of species in samples investigated in NMR experiments Zeolite + Adsorbate
Amount of acid sitesa (mmol/g)
βcitronellene (mmol/g)
Beta dry + β0.516 3.34 citronellene Beta wet + β0.516 3.34 citronellene Beta dry + β0.516 0.834 citronellene + Ethanol a. Measured by pyridine adsorption
Water (mmol/ g)
Ethanol (mmol/g )
0
0
6.1
0
0
8.34
Experiment 1 Liquid β-citronellene and dehydrated zeolite beta were mixed in a proportion of 0.6 ml/g. The added liquid volume of βcitronellene exceeded the zeolite micropore volume (0.23 ml/g). The 13C Hahn-echo NMR spectrum (Figure 3b red) exhibits the expected resonances of β-citronellene next to resonances due to isomers of the β-citronellene molecule formed catalytically during the NMR experiment (4 hours). The assignment of these resonances to individual isomers is provided in Supporting Information (S4). As mentioned above, direct acquisition of the 13C NMR spectrum through a Hahn-echo pulse sequence reveals resonances of mobile and adsorbed species. A broad underlying resonance extending from ca. 10 ppm to 50 ppm can be observed. It is ascribed to aliphatic C atoms of β-citronellene molecules (Figure 2) and the different isomers (Supporting Information S4), all with limited mobility, i.e. adsorbed in the zeolite pores. The narrow resonances of olefinic C atoms of β-citronellene and its isomers with their different chemical shifts were assigned (Figure 2 and Supporting information S4). In the 13C CPMAS NMR spectrum (Figure 3b black), only the rigid molecules are observed. This spectrum corresponds to the broad resonances that were observed on the Hahn-echo spectrum, which confirms the assignment to adsorbed molecules. These 13C resonances are very broad, reflecting the variety of conformations of βcitronellene and its isomers inside zeolite pores. Finally, narrow resonances of reagent and product molecules indicate that a fraction of the remaining β-citronellene and products are
ACS Paragon Plus Environment
Journal of the American Chemical Society
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 8
Figure 3. 13C{1H} solid-state MAS NMR spectra acquired with a Hahn-echo (in red) and CP (in black): (a) 13C liquid-state NMR spectrum; (b) dehydrated zeolite beta mixed with β-citronellene (Experiment 1); (c) hydrated zeolite beta mixed with β-citronellene (Experiment 2); CP spectrum magnified for clarity (in grey); (d) dehydrated zeolite beta mixed with β-citronellene and ethanol (molar ratio of 1 : 10); number of scans = 4096 (b, c & d). in liquid state, outside the zeolite. NMR experiment 1 confirms the penetration and reaction of β-citronellene in the zeolite pores at room temperature. Experiment 2 A sample of zeolite beta containing adsorbed moisture from ambient air (0.11 ml/g) was combined with liquid βcitronellene (0.6 ml/g) and filled in the NMR rotor. The 13C Hahn-Echo NMR spectrum (Figure 3c red) displays essentially narrow resonances, which reveal the presence of βcitronellene molecules in mobile state, i.e. outside the pores. The 13C CPMAS spectrum (Figure 3c grey) visualized the signature of 2,6-dimethyl-oct-7-en-2-ol characterized by resonances at ca. 75 ppm and ca. 41 ppm. This reaction product is formed via hydration of the β double bond of β-citronellene. The two resonances at 113 ppm and 145 ppm are assigned to the β-citronellene and α-double bond of the β-olefin hydration product (Figure 2). The limited line width of resonances of the CPMAS spectrum of the sample with hydrated zeolite beta (Figure 3c black) compared to dehydrated zeolite (Figure 3b black) is due to the uniqueness of the product formed. Comparison of the NMR data of β-citronellene adsorbed on dehydrated and hydrated zeolite beta (Figures 3b and c) therefore shows that the water molecules in the zeolite pores strongly compete with β-citronellene for adsorption. Experiment 3 The etherification reaction was performed by contacting dehydrated zeolite beta with a mixture of β-citronellene and ethanol. In the 13C Hahn-Echo NMR spectrum, βcitronellene C atoms appear as narrow resonances, while ethanol carbon atoms give rise to broadened 13C resonances at 57 ppm and 18 ppm (Figure 3d red). Ethanol is adsorbed in the zeolite pores with limited mobility, while the β-citronellene molecules exhibiting narrow signals exhibit a large degree of
freedom like in liquid-state. In the 13C CPMAS NMR spectrum (Figure 3d black), ethanol molecules remain characterized by broad peaks, but β-citronellene shows up with narrow resonances, as narrow as in the Hahn-echo spectrum. This last feature in CPMAS is ascribed to β-citronellene molecules in adsorbed state enabling residual dipolar coupling and polarization transfer while at the same time retaining a high degree of conformational freedom. This proves that the β-citronellene molecules are adsorbed at sites enabling mobility, which can be the pore entrances. It is difficult to envision another location in the sample which can provide such NMR characteristics. Citronellyl ether formation is indicated by the emergence of 13C resonances at 74.6 ppm, 56.6 ppm and 40.6 ppm (Figure 2). The earlier observed etherification selectivity at the βdouble bond compared to the α-double bond in β-citronellene23 is confirmed by the gradual disappearance of the 13C resonances of the β-double bond carbons (125 ppm and 130.9 ppm, Figure 2) while resonances of α-double bond carbons (112.7 ppm and ca. 144.8 ppm) are unaffected (Figure 4). The NMR resonances of the citronellyl ether exhibit narrow line widths in the CPMAS spectrum, similar to β-citronellene. The ether product visualized with CPMAS resides on the same location as the reagent, i.e. on pore mouths. This location of the molecules according to the NMR experiments offers insight into the reaction mechanism and explains the selective reaction. The occupation of the pores with ethanol forces βcitronellene molecules to react with ethanol at pore openings on the external surface. Isomerization reaction requiring penetration of β-citronellene deeper into the pores is suppressed. The high activity of zeolite beta in the etherification reaction is due to its 3D pore system and small crystallite size (Figure 1) offering plenty of readily available acid sites at pore mouths according to the high ACI value of 0.23 (Table 1). A possible reaction mechanism of etherification is presented in Scheme 1. The high selectivity for etherification at the β-double bond
ACS Paragon Plus Environment
Page 5 of 8
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Journal of the American Chemical Society
suggests the reaction proceeds by protonation of β-citronellene rather than ethanol. Etherification at the β-double bond is kinetically favored because of formation of a tertiary alkylcarbenium ion reaction intermediate on β-double bond protonation being more stable than the secondary alkylcarbenium ion intermediate formed upon protonation at the α-double bond.23 The stability of the alkylcarbenium ion can be critical when reaction occurs on weaker acid sites in pore mouths and in competitive adsorption with ethanol. Isomerization (and oligomerization leading to coke formation observed in kinetic experiments at 80-120°C)23 involving both α- and β-double bonds of β-citronellene (Supporting information) catalyzed by acid sites deeper inside the zeolite pores are suppressed by the limited access of the molecule to the pores. Many reactions catalyzed by zeolite beta involve water and/or alcohols.16–22 Competitive adsorption in favor of water and alcohol observed in this study offers an explanation why zeolite beta performs well in many types of liquid phase catalysis.
investigate the influence of external surface acid sites of MFI nanosheets in catalytic cracking of triisopropylbenzene.48 β-citronellene etherification was performed on zeolite beta catalyst in a fixed bed continuous flow liquid phase reactor using a feed mixture with a molar ratio of βcitronellene : EtOH of 1 : 10 at a pressure of 6 MPa. At a reaction temperature of 80°C, β-citronellene conversion reached ca. 50% with a selectivity to the ether of ca. 80%. Ca. 90% chemoselectivity for etherification at the β-double bond as compared to the α-double bond was achieved (Figure 5a). Addition of 1 wt% 2,4,6-collidine to the feed caused the activity of zeolite beta catalyst to decrease with increasing time on stream until 6 h, after which the catalyst was entirely poisoned (Figure 5a). The attenuation of catalytic activity by 2,4,6-collidine, a base molecule unable to penetrate the pores confirms that the catalytic sites for etherification are indeed located at the pore mouths. A similar poisoning experiment using 2,6-ditertiary butyl pyridine (kinetic diameter > 1.05 nm),49,50 showed a similar deactivation pattern (figure 5b). Parvulescu et al. investigated spent zeolite beta catalysts for etherification reactions of long linear olefins with alcohols like glycols and glycerol using confocal fluorescence spectroscopy.51 Based on the distribution of coke products on the zeolite crystals, they concluded that activity is concentrated at the rim of the crystals, like in this study.51
Figure 4. Time evolution of 13C{1H} solid-state MAS NMR Hahn-echo spectrum of zeolite beta contacted with mixture of β-citronellene and ethanol (molar ratio of 1:10) and 17 wt% nheptane as internal standard. Resonances assigned to citronellyl ether are indicated. The yellow frame highlights the disappearance of resonances of the β-double bond of β-citronellene with time.
Scheme 1. Proposed reaction scheme of β-citronellene etherification with ethanol on pore mouth acid sites of zeolite beta. Selective poisoning of pore mouth acid sites using bulky base molecules that do not enter the channels of the zeolite can be used to probe pore mouth catalysis in situ in a reactor. Du et al. used 2,4,6-collidine poisoning to demonstrate the role played by easily accessible acid sites in benzene alkylation with ethylene over acid MCM-22 zeolite.47 Kim et al. used selective poisoning with triphenyl phosphine oxide to
Figure 5. Time evolution of conversion (black trace), etherification yield (red trace) and isomer yield (blue trace) in βcitronellene etherification reaction with ethanol over H-beta zeolite (circle), 2,4,6-collidine poisoned H-beta zeolite (triangle, figure 5a) and 2,6-ditertiarybutyl pyridine poisoned Hbeta zeolite (inverted triangle, figure 5b). Reaction conditions: T = 80°C; Feed: 1 : 10 β-citronellene : EtOH (mol : mol), 17 wt% n-heptane and with / without ca. 1 wt% 2,4,6-Collidine or 2,6-ditertiarybutyl pyridine; WHSV = 3.8 h-1.
ACS Paragon Plus Environment
Journal of the American Chemical Society
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Conclusions The reaction mechanism of etherification of βcitronellene with ethanol in liquid phase on zeolite beta is revealed using in situ 13C Hahn-echo and CPMAS NMR spectroscopy. Undesired isomerization reactions of β-citronellene can be suppressed by preventing the molecule from penetrating the pores of the zeolite. This is achieved in presence of excess ethanol being preferentially adsorbed in the zeolite pores. The etherification reaction proceeds at easily accessible acid sites in pore openings. The β -double bond of βcitronellene preferentially is protonated by acid sites at such location leading to selective etherification reaction with ethanol molecule provided from inside the pores. Catalysis occurring at pore mouths is supported by the observed activity attenuation on selective acid site poisoning with bulky base molecules. Zeolite beta with its small crystal size (20-60 nm) is uniquely suited for liquid phase catalysis involving small molecules such as alcohols and water preventing penetration of larger organic reagents deep in the pores where they can undergo undesired side reactions and give rise to coke formation and deactivation. The here presented NMR approaches are also applicable to probe selective adsorption on zeolite in contact with multicomponent liquid. Supporting Information In situ FT-IR Investigation of pore mouth blocking on 2,4,6collidine adsorption; Equations for calculation of conversion and selectivity in catalytic reactions; Signal assignment of βcitronellene and citronellyl ethers using liquid state NMR spectroscopy; Identification of β-citronellene isomers formed in situ on impregnation of dry beta zeolite with β-citronellene; N2 sorption isotherm; FT-IR spectra of pyridine adsorbed on zeolite beta; FT-IR spectra of 2,4,6-collidine adsorbed on zeolite beta; 2,4,6-collidine penetration estimation.
AUTHOR INFORMATION Corresponding Author * Prof. Johan A. Martens, Centre for Surface Chemistry and Catalysis, KU Leuven, Celestijnenlaan 200F Bus 2461, 3001Heverlee, Leuven, Belgium. E-mail: Johan.Martens@biw.kuleuven.be; Fax: +32 1632 1998; Tel: +32 1632 1637 Present Addresses Dr. Pieter C. M. M. Magusin, The Grey group, Department of Chemistry, University of Cambridge, Lensfield Road, Cambridge, CB2 1EW Author Contributions All authors have given approval to the final version of the manuscript.
ABBREVIATIONS GC – Gas chromatography; NMR – Nuclear Magnetic Resonance; MAS – Magic angle spinning; CPMAS – Crosspolarization magic angle spinning NMR; ppm – parts per million; BET – Brunauer-Emmett-Teller method.
References (1)
(2)
(3)
(4)
(5) (6) (7)
(8) (9)
(10)
(11) (12) (13) (14)
(15)
Acknowledgements J.A.M. acknowledges the Flemish Government for long-term structural funding (Methusalem) and the Belgian government for interuniversity attraction poles (IAP). This work was supported by the Flemish FWO. CD acknowledges FWO– Vlanderen, BOF-UGent (GOA 01G01513) and the Hercules Foundation (AUGE/09/014) for funding. The authors acknowledge Karel Deurinckx for liquid state NMR measurements.
Page 6 of 8
(16) (17) (18)
(19)
Zeolites and Catalysis: Synthesis, Reactions and Applications; Cejka, J., Corma, A., Zones, S., Eds.; Wiley VCH: Weinheim 2010. Martens, J. A.; Jacobs, P. A. Some Aspects of Molecular Shape Selective Catalysis with Hydrocarbons in Zeolites. In Zeolite Microporous Solids: Synthesis, Structure, and Reactivity, NATO ASI Series Volume 352; Derouane, E. G., Lemos, F., Naccache, C., Ribeiro, F. R., Eds.; Springer Netherlands: Dordrecht, 1992; pp 511–529. Deactivation and Regeneration of Zeolite Catalysts - Catalytic Science Series: Volume 9; Guisnet, M., Ribeiro, F. R., Eds.; Imperial College Press: London, 2011. Venuto, P. B. Aromatic Reactions over Molecular Sieve Catalysts: A Mechanistic Review. In Catalysis in Organic Syntheses 1977; Gerard Smith, Ed.; Academic Press, New York, 1977; pp: 67-93. Fraenkel, D.; Cherniavsky, M.; Ittah, B.; Levy, M. J. Catal. 1986, 101, 273. Derouane, E. G.; Andre, J.-M.; Lucas, A. A. J. Catal. 1988, 110, 58. Martens, J. A.; Vanbutsele, G.; Jacobs, P. A.; Denayer, J.; Ocakoglu, R.; Baron, G.; Muñoz Arroyo, J. A.; Thybaut, J.; Marin, G. B. Catal. Today 2001, 65, 111. Claude, M. C.; Martens, J. A. J. Catal. 2000, 190, 39. Martens, J. A.; Verboekend, D.; Thomas, K.; Vanbutsele, G.; Pérez-Ramírez, J.; Gilson, J.-P. Catal. Today 2013, 218219, 135. Martens, J. A.; Souverijns, W.; Verrelst, W.; Parton, R.; Froment, G. F.; Jacobs, P. A. Angew. Chemie Int. Ed. English 1995, 34, 2528. Zecevic, J.; Vanbutsele, G.; de Jong, K. P.; Martens, J. A. Nature 2015, 528, 245. Ocakoglu, R. A.; Denayer, J. F. M.; Marin, G. B.; Martens, J. A.; Baron, G. V. J. Phys. Chem. B 2003, 107, 398. Katzer, J. R., Ph. D. thesis, Massachusetts Institute of Technology, Cambridge, Mass., 1969. Laxmi Narasimhan, C. S.; Thybaut, J. W.; Marin, G. B.; Jacobs, P. A.; Martens, J. A.; Denayer, J. F.; Baron, G. V. J. Catal. 2003, 220, 399. Laxmi Narasimhan, C. S.; Thybaut, J. W.; Marin, G. B.; Denayer, J. F.; Baron, G. V.; Martens, J. A.; Jacobs, P. A. Chem. Eng. Sci. 2004, 59, 4765. Goodwin, J. G.; Natesakhawat, S.; Nikolopoulos, A. A.; Kim, S. Y. Catal. Rev. Sci. Eng. 2002, 44, 287. Jansen, J. C.; Creyghton, E. J.; Njo, S. L.; van Koningsveld, H. and; van Bekkum, H. Catal. Today 1997, 38, 205. Calvin Mukarakate, Michael J. Watson, Jeroen ten Dam, Xavier Baucherel, B.; Sridhar Budhi, Matthew M. Yung, Haoxi Ben, Kristiina Iisa, R. M. B. and; Nimlos, M. R. Green Chem. 2014, 1433, 423. Aho, A.; Kumar, N.; Eränen, K.; Salmi, T.; Hupa, M.; Mur-
ACS Paragon Plus Environment
Page 7 of 8
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
(20)
(21) (22) (23)
(24) (25)
(26) (27) (28) (29) (30)
(31) (32) (33)
(34) (35)
(36) (37)
(38) (39) (40) (41) (42) (43) (44)
(45)
Journal of the American Chemical Society zin, D. Y. Process Saf. Environ. Prot. 2007, 85, 473. Dijkmans, J.; Dusselier, M.; Gabriëls, D.; Houthoofd, K.; Magusin, P. C. M. M.; Huang, S.; Pontikes, Y.; Trekels, M.; Vantomme, A.; Giebeler, L.; Oswald, S.; Sels, B. F. ACS Catal. 2015, 5, 928. Moliner, M.; Roman-Leshkov, Y.; Davis, M. E. Proc. Natl. Acad. Sci. 2010, 107, 6164. Corma, A.; Nemeth, L. T.; Renz, M.; Valencia, S. Nature 2001, 412, 423. Radhakrishnan, S.; Purino, M.; Alexopoulos, K.; Taulelle, F.; Reyniers, M.-F.; Marin, G. B.; Martens, J. A. Green Chem. 2015, 17, 2840. Radhakrishnan, S.; Franken, J.; Martens, J. A. Green Chem. 2012, 14, 1475. Radhakrishnan, S.; Thoelen, G.; Franken, J.; Degrève, J.; Kirschhock, C. E. A.; Martens, J. A. ChemCatChem 2013, 5, 576. Ivanova, I. I.; Kolyagin, Y. G. Chem. Soc. Rev. 2010, 39, 5018. Haw, J. F.; Song, W.; Marcus, D. M.; Nicholas, J. B. Acc. Chem. Res. 2003, 36, 317. Wang, W.; Jiang, Y.; Hunger, M. Catal. Today 2006, 113, 102. Haouas, M.; Walspurger, S.; Taulelle, F.; Sommer, J. J. Am. Chem. Soc. 2004, 126, 599. Wolf, P.; Valla, M.; Rossini, A. J.; Comas-Vives, A.; Núñez-Zarur, F.; Malaman, B.; Lesage, A.; Emsley, L.; Copéret, C.; Hermans, I. Angew. Chem. Int. Ed. Engl. 2014, 53, 10179. Borghèse, S.; Haouas, M.; Sommer, J.; Taulelle, F. J. Catal. 2013, 305, 130. Haouas, M.; Fink, G.; Taulelle, F.; Sommer, J. Chemistry 2010, 16, 9034. Arzumanov, S. S.; Reshetnikov, S. I.; Stepanov, A. G.; Parmon, V. N.; Freude, D. J. Phys. Chem. B 2005, 109, 19748. Stepanov, A. G.; Sidelnikov, V. N.; Zamaraev, K. I. Chem. A Eur. J. 1996, 2, 157. Gabrienko, A. A.; Arzumanov, S. S.; Luzgin, M. V.; Stepanov, A. G.; Parmon, V. N. J. Phys. Chem. C 2015, 119, 24910. Emeis, C. A. J. Catal. 1993, 141, 347. Nesterenko, N. S.; Thibault-Starzyk, F.; Montouilliout, V.; Yushchenko, V. V.; Fernandez, C.; Gilson, J.-P.; Fajula, F.; Ivanova, I. I. Kinet. Catal. 2006, 47, 40. Hahn, E., Phys. Rev. 1950, 80, 580-594. Pines, A.; Gibby, M.G.; Waugh, J.S. Chem. Phys. Lett. 1975, 15, 373-376. Hartmann, S.; Hahn. E; Phys. Rev. 1962, 128, 2042-2053. Fung, B.; Khitrin, A., Ermolaev, K. J. Magn. Res. 2000, 142, 97–101. Dietz, W. A. J. Chromatogr. Sci. 1967, 5, 68. Nanda, M. R.; Yuan, Z.; Qin, W.; Ghaziaskar, H. S.; Poirier, M.-A.; Xu, C. (Charles). Appl. Energy 2014, 123, 75. International Zeolite Association: Database of Zeolite Structures - Framework Type *BEA http://izasc.biw.kuleuven.be/fmi/xsl/IZA-SC/ftc_fw.xsl?db=Atlas_main&-lay=fw&-max=25&STC=BEA&-find (accessed: January 29, 2016). Thibault-Starzyk, F.; Stan, I.; Abelló, S.; Bonilla, A.;
(46)
(47) (48) (49) (50) (51)
Thomas, K.; Fernandez, C.; Gilson, J.-P.; Pérez-Ramírez, J. J. Catal. 2009, 264, 11. Kerkhofs, S.; Saïdi, F.; Vandervoort, N.; Van den Mooter, G.; Martineau, C.; Taulelle, F.; Martens, J. A. J. Mater. Chem. B 2015, 3, 3054. Du, H.; Olson, D. H. J. Phys. Chem. B 2002, 106, 395. Kim, W.; Ryoo, R. Catal. Letters 2014, 144, 1164. Góra-Marek, K.; Tarach, K.; Choi, M. J. Phys. Chem. C 2014, 118, 12266. Fǎrcaşiu, D.; Leu, R.; Corma, A. J. Phys. Chem. B 2002, 106, 928. Parvulescu, A. N.; Mores, D.; Stavitski, E.; Teodorescu, C. M.; Bruijnincx, P. C. A.; Gebbink, R. J. M. K.; Weckhuysen, B. M. J. Am. Chem. Soc. 2010, 132, 10429.
ACS Paragon Plus Environment
Journal of the American Chemical Society
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 ACS Paragon Plus Environment
Page 8 of 8